Original Article

Clinical Pharmacology
A Multiple‐Dose, Randomized, Double‐ in Drug Development
2(3) 285–294

Blind, Placebo‐Controlled, Parallel‐Group © The Author(s) 2013

DOI: 10.1002/cpdd.36

QT/QTc Study to Evaluate the

Electrophysiologic Effects of THC/CBD
Spray

Edward M. Sellers1, Kerri Schoedel2, Cindy Bartlett2, Myroslava Romach1,

Ethan B. Russo3, Colin G. Stott3, Stephen Wright3, Linda White3,
Paul Duncombe3, and Chien‐Feng Chen4

Abstract
Delta‐9‐tetrahydrocannabinol (THC)/cannabidiol (CBD) oromucosal spray has proved efficacious in the treatment of
spasticity in multiple sclerosis and chronic pain. A thorough QT/QTc study was performed to investigate the effects of
THC/CBD spray on electrocardiogram (ECG) parameters in compliance with regulatory requirements, evaluating the
effect of a recommended daily dose (8 sprays/day) and supratherapeutic doses (24 or 36 sprays/day) of THC/CBD spray
on the QT/QTc interval in 258 healthy volunteers. The safety, tolerability, and pharmacokinetic profile of THC/CBD
spray were also evaluated. Therapeutic and supratherapeutic doses of THC/CBD spray had no effect on cardiac
repolarization with primary and secondary endpoints of QTcI and QTcF/QTcB, respectively, showing similar results.
There was no indication of any effect on heart rate, atrioventricular conduction, or cardiac depolarization and no new
clinically relevant morphological changes were observed. Overall, 19 subjects (25.0%) in the supratherapeutic (24/36 daily
sprays of THC/CBD spray) dose group and one (1.6%) in the moxifloxacin group withdrew early due to intolerable AEs.
Four psychiatric serious adverse events (AEs) in the highest dose group resulted in a reduction in the surpatherapeutic
dose to 24 sprays/day. In conclusion, THC/CBD spray does not significantly affect ECG parameters. Additionally, THC/
CBD spray is well tolerated at therapeutic doses with an AE profile similar to previous clinical studies.

Keywords
cannabinoids, tetrahydrocannabinol, cannabidiol, cardiotoxicity, toxic psychosis

An undesirable property of some non-antiarrhythmic
drugs is their ability to delay cardiac repolarization,
which puts patients at risk of sudden cardiac death.
Current regulatory guidance emphasizes the need for
clear robust data on the effect of new chemical entities on
electrocardiogram (ECG) parameters,1 with focus on
cardiac repolarization, as measured by the QTc duration.
Delta-9-tetrahydrocannabinol (THC)/cannabidiol
(CBD) spray (Sativex) is an oromucosal spray containing
two principal cannabinoids, THC and CBD. These
extracts are prepared from the Cannabis sativa L. plant
and formulated into an approximate 1:1 ratio alongside
minor cannabinoids and other plant components, includ-
ing but not limited to the phytocannabinoids cannabigerol
and cannaibicromene, as well as cannabis terpenoids.2
There is much interest in the future development of THC/
CBD spray as it has promising efficacy in several
therapeutic areas and currently has approval for use in the
UK, Spain, Czech Republic, New Zealand, Germany,
Sweden, Denmark, and Canada for the treatment of
spasticity due to multiple sclerosis (MS), and is also
approved and marketed in Canada for the relief of
neuropathic pain in MS and as an adjunctive treatment in
1
DL Global Partners, Inc., Toronto, ON, Canada
2
INC Research, Toronto, Toronto, ON, Canada
3
GW Pharma Ltd, Salisbury, Wiltshire, UK
4
Otsuka Pharmaceutical Development and Commercialization, Inc.,
Princeton, NJ, USA
Submitted for publication 25 September 2012; accepted 8 March 2013
Corresponding Author:
Dr Edward M. Sellers, DL Global Partners, Inc., 78 Baby Point Cres,
Toronto, ON M6S 2C1, Canada
(e‐mail: e.sellers@dlglobalpartners.com)
286
cancer pain. THC/CBD spray has also shown positive
effects on non-MS neuropathic pain,3,4 rheumatoid
arthritis,5 and bladder dysfunction.6
Cannabis use may be associated with tachycardia and
transient changes in blood pressure.7–11 There was also
one report of antiarrhythmic effects after large doses in
subjects treated with intravenous THC.12 Although there
are few reports of cardiovascular events associated with
marijuana use, it is feasible that the real incidence of
arrhythmias is substantially underreported given the
prohibition of cannabis use.13 Despite this, a large cohort
study showed no association between marijuana use and
hospitalization or mortality associated with cardiovascu-
lar disease.14
Given the regulatory requirements and the potential
cardiovascular effects of cannabinoids, it was felt
necessary to evaluate the effects of THC/CBD spray on
the QT/QTc interval in humans. As such, the effects of
THC/CBD spray on ECG parameters were investigated.
Methods
This multiple-dose, randomized, double-blind, double-
dummy, placebo- and active-controlled, four-arm, paral-
lel-group study, designed to evaluate the effect of THC/
CBD spray on the QT/QTc interval, was performed in 258
healthy male and female subjects aged between 18 and
45 years inclusive. It was carried out by an independent
contract research organization at one study center
(DecisionLine Clinical Research Corporation, ON,
Canada), with all subjects enrolled on-site. It was
designed in accordance with the Declaration of Helsinki,
and with adherence to the principles of Good Clinical
Practice (GCP), outlined by the ICH GCP guidelines, was
reviewed and approved by the Ontario Institutional
Review Board, and all participants provided written
informed consent.
Primary Objective
The primary objective was to evaluate the effects of
therapeutic and supratherapeutic doses of THC/CBD
spray on ECG parameters.
Secondary Objectives
Secondary objectives were to evaluate the safety,
tolerability, and pharmacokinetics (PKs; CBD, THC
and its metabolite 11-hydroxy-THC [11-OH-THC]) of
THC/CBD spray following multiple dose administration,
and to evaluate the relationship between the PKs and the
QT/QTc interval.
Inclusion Criteria
Eligible subjects were English speakers aged 18–45 with
a body mass index (BMI) between 19 and 29.9 kg/m2, and
a minimum weight of 50 kg at screening. Eligible subjects
Clinical Pharmacology in Drug Development 2(3)
were non-tobacco users within 6 months of study onset,
were free from clinically significant abnormalities, and
had systolic and diastolic blood pressures between 90–
140 and 50–90 mmHg at screening, respectively. Those
of reproductive potential were required to ensure
effective contraception was used within 1 month of study
onset, for the study duration, and for a minimum of
60 days following study completion. Female subjects
were required to have a negative pregnancy test result,
were to not be lactating or planning a pregnancy for at
least 60 days following study completion. All subjects
also had to be willing to abide by all study requirements
and restrictions.
Exclusion Criteria
Subjects with a history of significant psychiatric, renal,
hepatic, gastrointestinal, endocrine, immunologic, der-
matologic, oncologic, cardiovascular, or convulsive
disorder or with a known hypersensitivity to the study
medication were excluded. Subjects taking a non-
prescription medication (within 7 days of study onset),
a prescription, natural, or recreational drug (within
14 days of first dosing and throughout the study), or
those who had received any investigational medicinal
product within 30 days of screening were excluded, as
were those with a known history of alcohol or substance
abuse. Those using cannabis or a THC-containing
medicine within 3 months of study entry were excluded,
as were those with habituation to analgesic drugs and
those unable to refrain from consuming caffeine for at
least 9 days. Those who were positive for Hepatitis B,
Hepatitis C, or HIV at screening were excluded, as were
those with current or pending legal charges, or on
probation. Subjects who had donated more than 50 mL of
blood within 30 days, or more than 400 mL with 56 days
of admission for treatment were excluded, as were those
who were study site employees, or relatives of employees
directly involved in the study.
Treatment Groups and Dosing
Subjects attended a screening visit to determine
eligibility, then, within 30 days were admitted to the
study site 2 days prior to Dosing (Day
1) to reassess
eligibility. The following day (Day 0) triplicate (i.e., three
simultaneous recordings) baseline ECG measurements
were performed at 14 specified times (0.5, 1, 1.5, 2, 2.5, 3,
4, 6, 8, 10, 12, 16, 20, and 23.5 hours post the time of the
planned first dose on Days 1–5) to provide averaged time-
matched baseline QT intervals for comparison with the
intervals obtained later in the treatment session.
Dosing. Each 100 mL spray of THC/CBD spray
contained 2.7 mg THC and 2.5 mg CBD. The therapeutic
dose of THC/CBD spray chosen for the study was eight
daily sprays which totaled 21.6 mg THC þ 20 mg CBD,
and the supratherapeutic doses of 24 or 36 daily sprays
Sellers et al
totaled 64.8 mg THC þ 60 mg CBD (24 sprays) or
97.2 mg THC þ 90 mg CBD (36 sprays), respectively
(36 daily sprays was the initial supratherapeutic dose, but
this was later reduced to 24 daily sprays due to four
subjects experiencing psychiatric serious adverse events
[SAEs]). Placebo spray contained the colourants plus
excipients, and placebo tablets were similar in size, shape,
and color to moxifloxacin tablets but contained no active
ingredient. Moxifloxacin tablets (400 mg) served as the
positive control.
Subjects were randomized to one of four treatment
groups according to a pre-generated randomization
schedule:
Group 1: received 24 or 36 placebo sprays
(divided into 12 or 18 sprays twice daily,
12 hours apart) for 5 days and a single oral
moxifloxacin placebo on Day 5.
Group 2: received eight sprays of THC/CBD
spray (four sprays twice daily, 12 hours apart)
and 16 or 28 placebo sprays (eight or 14 sprays
twice daily every 12 hours) for 5 days and a
single oral moxifloxacin placebo on the
morning of Day 5.
Group 3: received 24 or 36 THC/CBD sprays
(12 or 18 sprays twice daily 12 hours apart) for
5 days and a single oral moxifloxacin placebo
on the morning of Day 5.
 CBD spray dose and moxifloxacin. As such, a parallel-
It should be noted that during the study, the
supratherapeutic dose was reduced from 36 sprays
to 24 sprays THC/CBD spray following SAEs in
four subjects (in 162 exposures, 2.5%) who
received THC/CBD sprays at 36 sprays. ECG
data were analyzed for the two supratherapeutic
doses combined (n ¼ 52 total; 29 of these
subjects received THC/CBD spray at 24 sprays,
and 23 subjects received THC/CBD spray at 36
sprays).
Group 4: received 24 or 36 placebo sprays (12 or
18 sprays twice daily 12 hours apart) for 5 days
and a single oral moxifloxacin 400 mg tablet
on the morning of Day 5.
Blinding was maintained by using two vials of spray.
At each dose, four THC/CBD spray or placebo sprays
were administered from Vial 1 and the remaining eight
THC/CBD spray or placebo sprays were administered
from Vial 2.
THC/CBD spray or placebo administration was
started on Day 1 and continued for 5 days. On
the morning of Day 5, subjects received either moxi-
floxacin 400 mg (Group 4) or moxifloxacin placebo
(Groups 1, 2, and 3) in addition to the THC/CBD spray/
placebo sprays.
287
On Days 1–4, vital signs and safety ECG measure-
ments were collected pre-dose and at 2 hours post-dose.
Triplicate ECG measurements, PK samples and vital
signs measurements were collected on Day 5 at exactly
the same time points as the ECG measurements taken on
Day 0. In addition, a safety EGC was performed pre- and
approximately 2 hours post-dose.
Adverse events (AEs) were recorded throughout the
study period. The last post-dose procedures were
completed 23.5 hours after the last dose and then a final
safety follow-up visit was performed 7–14 days later.
Study Design
The study was designed to take into account the long
terminal elimination half-life of THC/CBD spray and the
potential for lower tolerability of THC/CBD spray in
healthy, drug naı̈ve subjects.
Despite sampling of plasma concentration levels
during chronic THC/CBD spray dosing suggesting that
significant accumulation does not occur, and although a
crossover design is often recommended for these studies,1
it was not appropriate in this instance for several reasons.
Firstly, the distribution and slow release of THC and
active metabolites from adipose tissue results in a
prolonged terminal elimination half-life. Secondly, since
high doses of THC/CBD spray were used, a crossover
design would increase the risk of subject withdrawals,
which would hinder data collection for the lower THC/
group multiple-dose trial design was selected to evaluate
the effects of THC/CBD spray on QT/QTc intervals.
In accordance with ICH guidelines moxifloxacin
tablets (400 mg) served as the positive control to establish
assay sensitivity since it is known to have an effect on the
mean QT/QTc interval of about 5 milliseconds (ms).1 The
study was performed in healthy volunteers to eliminate
variables known to change ECG parameters such as
concomitant drugs, diseases etc. All subjects were either
cannabis-naı̈ve or had not taken cannabis or THC-
containing medicines for at least 3 months prior to study
entry.
Assay for the Measurement of Moxifloxacin
Assay method. The assay was developed and validated
by Quotient Bioresearch Limited (Cambridge, UK), who
measured moxifloxacin in human plasma (100 flL) by
liquid chromatography-tandem mass spectrometry (LC–
MS/MS) following protein precipitation and reference to
a calibration line over the range 50–5,000 ng/mL,
extracted with each batch. Quality control (QC) samples
at three concentrations were interspersed with the test
samples at 150, 1,000, and 4,000 ng/mL for moxifloxacin.
The internal standard used in the assay was lomefloxacin.
Data processing. Data collection and peak area
integration were performed using Analyst software
288
(versions 1.4.1 and 1.4.2) associated with the mass
spectrometer. Standard regression and quantification
were performed using Watson L1MS (version 7.0). The
mass spectrometer response (peak area ratio of analyte
to internal standard) of each calibration standard was
plotted against the theoretical (prepared) concentra-
tions. This plot was subjected to least squares
regression analysis using a linear fit (weighted 1/x2)
to provide values for correlation coefficient and back-
calculated concentrations.
Acceptability of analytical batches. The acceptability of
each batch of test samples depended on the data from the
calibration standards and the QC samples fulfilling the
following requirements:
 At least 75% of calibration samples being
within 15% (20% at the Lower Limit of
Quantification [LLOQ]) of their target
concentration.
 At least four of the six QC samples being
within 15% of their respective target
values.
 Two of the six QC samples may be outside the
15% limit but not at the same concentration.
Assay for the Measurement of THC, 11‐OH‐THC,
and CBD
The assay was developed and validated by Quotient
Bioresearch Limited. Each study sample was initially
analyzed using an assay with a LLOQ of 0.5 ng/mL for
THC, 11-OH-THC, and CBD. Following the first phase
of analysis, a more sensitive analytical method with a
LLOQ of 0.1 ng/mL was developed and was used to
reanalyze specific samples which were originally Below
the Limit of Quantification.
Original assay (LLOQ of 0.5 ng/mL) method. Human
plasma samples (200 mL) were analyzed for THC, 11-
OH-THC, and CBD by LC–MS/MS after supported
liquid extraction, and referenced to a calibration line
over the range 0.5–100 ng/mL for THC, 11-OH-THC,
and CBD, extracted with each batch. QC samples at
three concentrations were interspersed with the test
samples at 1.5, 40, and 75 ng/mL for THC, 11-OH-
THC, and CBD. The respective internal standards
used in the assay were d3-THC, d3-11-OH-THC, and
d3-CBD.
More sensitive assay (LLOQ of 0.1 ng/mL) method. Hu-
man plasma samples (500 mL) were analyzed for THC,
11-OH-THC, and CBD by LC–MS/MS after solid-phase
extraction and referenced to a calibration line over the
range 0.1–100 ng/mL for THC and 0.1–30 ng/mL for 11-
OH-THC and CBD, extracted with each batch. QC
samples at three concentrations were interspersed with
the test samples at 0.3, 45, and 80 ng/mL for THC, and at
Clinical Pharmacology in Drug Development 2(3)
0.3, 15, and 23 ng/mL for 11-OH-THC and CBD. The
respective internal standards used in the assay were d3-
THC, d3-11-OH-THC, and d3-CBD.
Analysis sequence. Each batch of samples was injected
as follows: serum separator tube, wash solvent, blank,
zero, calibration standards, extracted carryover blank,
unknown samples interspersed with QC samples, dupli-
cate calibration standards at LLOQ, and Upper Limit of
Quantification levels. Each batch consisted of 96 samples
or less.
Data Processing and acceptability of analytical batches.
Data processing was carried out as for moxifloxacin,
and the acceptability of analytical batches was also the
same.
Statistical Considerations
Sample size. The sample size was based on the non-
inferiority of THC/CBD spray against placebo in the
primary analysis. A sample size of 60 subjects per group
was selected to provide at least 80% power to show that
the upper limit of the 90% confidence interval (CI; two-
sided) for the comparison of THC/CBD spray and
placebo would fall below 10 ms.15
Triplicate ECG analysis. The primary analysis was a
time-matched analysis performed at all 14 time points to
investigate whether anyone had a placebo corrected mean
change from baseline in QTcI in which the upper two-
sided 90% (i.e., one-sided upper 95%) CI exceeded
10 ms. Placebo correction was calculated as the differ-
ence from the mean change of the same time point in the
subjects on placebo.
Change from mean of all baseline ECGs to the mean
of all on-treatment ECG values for a given subject for
each of: heart rate and PR, QRS, QT, QTc intervals
were calculated. Time-averaged analysis (i.e., mean of
all time points at baseline and steady state [Day 5]) was
also included to provide a context for other historical
drug data. New ECG morphological changes (present
on treatment but not baseline ECGs) were also
evaluated.
Correction formulae. QTcI was the individually deter-
mined QT correction, and the goal was to find b such that
QTcI was a constant, where QTcI ¼ QT/(RR)b where RR
is the interval from the onset of one QRS complex to the
onset of the next QRS complex.
Additional correction formulae that were included but
considered secondary were QTcF and QTcB, and were
defined as:
 QTcF was the length of the QT interval
corrected for heart rate by Fridericia’s
formula:
QT
QTcF ¼
ðRRÞ1=3
Sellers et al
 QTcB was the length of the QT interval
corrected for heart rate by Bazett’s formula:
QT
QTcB ¼
ðRRÞ1=2
Pharmacokinetics. PKs for CBD, THC and its metabo-
lite 11-OH-THC included Cmax, Tmax, AUC0–t, and
AUC0–inf (derived using non-compartmental data
analysis).
Methods Employed to Control for Variability
To control for variability in the study, a number of
methods were employed. Inclusion into the study was
limited to those aged between 18 and 45 due to the
potential association between increasing age and QTc
prolongation. The BMI of subjects enrolled was also
limited to between 19 and 29.9 kg/m2, since increasing
BMI has been associated with increases in the QTc
interval in healthy subjects, potentially leading to
ventricular hypertrophy and myocardial action potential
prolongation.16 Further measures to reduce variability in
the results included blinding ECG analysts and cardiol-
ogists to the treatment allocation, and having the same
trained analyst reading individual subjects’ ECGs, with
interval durations further verified by a cardiologist.
Results
Subject Disposition and Demographics
A total of 258 eligible subjects were randomized with 257
receiving at least one dose of study drug and 229
completing the study. Of the subjects randomized to each
treatment group, 98.3% for placebo (59 of 60 subjects),
98.4% for THC/CBD 8 sprays (60 of 61 subjects), and
98.4% for moxifloxacin (60 of 61 subjects) completed the
study. A total of 65.8% of subjects (50 of 76 subjects)
completed the study in the combined THC/CBD 24/36
sprays group: 80.0% (28 of 35 subjects) in the THC/CBD
24 sprays group and 53.7% (22 of 41 subjects) in the
THC/CBD 36 sprays group. Twenty-nine subjects
(11.2%) withdrew from the study prior to completion;
one subject each in the placebo, THC/CBD spray 8
sprays, and moxifloxacin groups (1.7%, 1.6%, and 1.6%,
respectively) and 26 subjects (34.2%) in the combined
THC/CBD spray 24/36 sprays group (seven subjects
taking 24 sprays and 19 subjects taking 36 sprays). The
subjects in the placebo and THC/CBD spray at eight
sprays groups withdrew consent, the subject in the
moxifloxacin group and 19 subjects in the THC/CBD
spray 24/36 sprays group withdrew due to intolerable
AEs, and the remaining seven subjects in the THC/CBD
spray 24/36 sprays groups withdrew consent. No subjects
discontinued due to marked prolongation of QT/QTc
289
interval alone. Two subjects were withdrawn due to AEs
of QT interval prolonged in addition to other AEs and
therefore were counted in the category of discontinua-
tions due to intolerable AEs rather than marked
prolongation of the QT/QTc interval. Both subjects
were in the 36 spray dose group. One subject experienced
toxic psychosis which resolved upon treatment cessation
and led to their withdrawal from the study, and the other
experienced intolerable AEs of mild abdominal pain
upper, dizziness, anxiety, fatigue, left upper quadrant
pain, and vomiting, with led to their withdrawal from the
study (their mild AE of prolonged QT resolved within
7 hours).
Baseline medical history and physical and oral
examinations revealed little of clinical significance.
There was a greater proportion of male (64.2%) to
female (35.8%) subjects overall in the study, with no
marked differences between treatment groups. The
average age of study participants was approximately
32 years, again with little variation between treatment
groups. The ethnic origins of subjects varied overall, but
were similar between groups.
Primary Objective Results
Triplicate ECG results. The time-matched analysis for
the QTcI endpoint revealed that assay sensitivity criteria
were met (Fig. 1). The placebo-corrected, time-matched
change in QTcI from baseline for moxifloxacin was
greater than 5 ms at all 14 time points (range of 6.1–
18.2 ms), including those time points pre-specified for
determination of assay sensitivity (1, 1.5, 2, 2.5, 3, and
4 hours), where differences ranged from 14.8 to 18.2 ms.
All upper bounds of the 90% CIs exceeded 10 ms
(10.4 ms) and those at the pre-specified time points
ranged from 19.1 to 22.5 ms.
For THC/CBD spray, most placebo-corrected
changes from baseline values were less than 5 ms,
ranging from
4.1 to 5.2 ms for THC/CBD spray 8
sprays and from
5.5 to 6.2 ms for THC/CBD spray
24/36 sprays. The upper bounds of the 90% CIs for
both THC/CBD spray treatment groups were below the
10 ms thresholds at all time-points, indicating no effect
on cardiac repolarization.
Time‐averaged analysis. Assay sensitivity was con-
firmed by the time-averaged analysis: placebo-corrected
change from baseline in QTcI for moxifloxacin exceeded
10 ms. The time-averaged data confirmed the lack of QTc
effect of THC/CBD spray. In both THC/CBD spray
groups, the upper bounds of the 90% CIs for placebo-
corrected change from baseline were less than 10 ms for
QTcI. Small non-significant changes observed in the
primary endpoint of time-matched, placebo-corrected
change in QTcI from baseline were slightly greater in the
therapeutic compared with the supratherapeutic THC/
CBD spray dose group.
290
Figure 1. Placebo‐corrected, change from baseline in QTcI (ms)—Estimated from mixedmodel analysis of variance (90% CI).
Secondary Objective Results
When the additional corrected QTc values of QTcF and
QTcB were calculated according to their appropriate
formulae, they showed similar results to the primary
endpoint of QTcI; there was no signal of a THC/CBD
spray effect on cardiac repolarization based on QTcF
(data not shown) and only a very small effect when based
on the less accurate QTcB (data not shown). Time-
averaged analysis for QTcF and QTcB mirrored the
finding for QTcI, further supporting the fact that THC/
CBD spray had no effect on cardiac repolarization (data
not shown).
Analysis of the remaining secondary endpoints of
heart rate, PR interval, QRS interval, and uncorrected QT
interval all showed little of note between treatment
groups.
Treatment emergent abnormalities in ECG morpho-
logical patterns observed on Day 5 included inverted T
wave abnormalities in four (7%) of subjects receiving
THC/CBD spray in the 8 spray group, three (5%)
receiving moxifloxacin, and three (5%) placebo subjects.
Other events included ST segment depression in two (4%)
of subjects in the 24/36 spray THC/CBD spray group, two
(3%) in the moxifloxacin group, and one (2%) in the
placebo group. None of the new morphological changes
observed was considered clinically significant.
Clinical Pharmacology in Drug Development 2(3)
Safety variables. AEs were analyzed separately for
Days 1–5 pre-dose and Day 5 post-dose onwards, because
the subjects in the moxifloxacin group received placebo
sprays from Day 1 to Day 4 and did not receive
moxifloxacin until Day 5. A summary of treatment-
emergent AEs which occurred in at least 10% of subjects
in any given treatment group is presented in Table 1.
Overall, the therapeutic dose of THC/CBD spray
(eight sprays) was well-tolerated, while the suprather-
apeutic doses (24 and 36 sprays) were not well-tolerated
with numerous central nervous system (CNS) and
psychiatric AEs reported. The most common AEs in
the study were somnolence, dizziness, euphoric mood,
headache, nausea, and dry mouth, and the most
commonly affected body systems were the nervous
system and gastrointestinal system, followed by psychi-
atric disorders and general disorders and administration
site conditions.
There were no deaths in this study. Four subjects
(9.8%) in the THC/CBD spray 36 sprays group
experienced SAEs, all of which were psychiatric in
nature. Most events had resolved within an hour
following treatment although one delusional disorder
took 3 days to resolve completely.
There were few apparent treatment- or dose-related
changes in laboratory parameters, although five subjects
Sellers et al 291

Table 1. Treatment‐Emergent Adverse Events (Preferred Term) Reported in at Least 10% of Subjects in Any Treatment Group

THC/CBD spray
Preferred Term Placebo Moxifloxacin
8 Sprays 24 Sprays 36 Sprays

Number of subjects (%)

1–5 5þ 1–5 5þ 1–5 5þ 1–5 5þ 1–5b 5þ

Day a
(n ¼ 60) (n ¼ 59) (n ¼ 60) (n ¼ 60) (n ¼ 35) (n ¼ 29) (n ¼ 41) (n ¼ 22) (n ¼ 61) (n ¼ 60)
Somnolence 7 (11.7) 6 (10.2) 16 (26.7) 6 (10.0) 12 (34.3) 6 (20.7) 20 (48.8) 4 (18.2) 12 (19.7) 1 (1.7)
Dizziness 3 (5.0) 0 18 (30.0) 4 (6.7) 19 (54.3) 2 (6.9) 17 (41.5) 2 (9.1) 4 (6.6) 3 (5.0)
Euphoric mood 4 (6.7) 0 10 (16.7) 2 (3.3) 15 (42.9) 1 (3.4) 16 (39.0) 0 0 1 (1.7)
Headache 10 (16.7) 4 (6.8) 6 (10.0) 3 (5.0) 5 (14.3) 3 (10.3) 7 (17.1) 0 3 (4.9) 4 (6.7)
Nausea 2 (3.3) 1 (1.7) 3 (5.0) 0 10 (28.6) 0 9 (22.0) 1 (4.5) 4 (6.6) 5 (8.3)
Dry mouth 0 0 6 (10.0) 0 12 (34.3) 2 (6.9) 3 (7.3) 0 1 (1.6) 0
Vomiting 1 (1.7) 0 1 (1.7) 0 5 (14.3) 0 10 (24.4) 1 (4.5) 2 (3.3) 2 (3.3)
Constipation 3 (5.0) 0 8 (13.3) 1 (1.7) 2 (5.7) 0 2 (4.9) 1 (4.5) 2 (3.3) 0
Fatigue 2 (3.3) 0 7 (11.7) 0 2 (5.7) 0 5 (12.2) 0 0 1 (1.7)
Pallor 0 0 1 (1.7) 1 (1.7) 4 (11.4) 0 5 (12.2) 0 0 0
Decreased appetite 0 0 0 0 2 (5.7) 0 5 (12.2) 1 (4.5) 1 (1.6) 1 (1.7)
Tachycardia 0 0 0 0 4 (11.4) 0 5 (12.2) 0 0 0
Feeling hot 0 0 0 1 (1.7) 1 (2.9) 0 5 (12.2) 1 (4.5) 0 0
Feeling of relaxation 0 0 2 (3.3) 0 4 (11.4) 0 2 (4.9) 0 2 (3.3) 0
Inappropriate affect 0 0 2 (3.3) 0 4 (11.4) 0 2 (4.9) 0 0 0
Disorientation 0 0 0 1 (1.7) 0 0 6 (14.6) 0 0 0
Days 1–5 ¼ Day 1 post‐dose to Day 5 pre‐dose; Day 5þ ¼ Day 5 post‐dose to end of study.
a
b
Moxifloxacin group received placebo on Day 1 to Day 5 pre‐dose.

had clinically significant abnormal liver function tests
that were recorded as AEs at follow-up. One subject in the
eight sprays THC/CBD spray group had a mild AE of
liver function tests abnormal (ALT; Alanine aminotrans-
ferase) at follow-up on study Day 15. Two subjects in the
24 sprays THC/CBD spray group had clinically signifi-
cant elevations in aspartate aminotransferase (AST) and
ALT values at follow-up, which were recorded as AEs of
hepatic enzyme increased. One subject in the 36 sprays
THC/CBD spray group had a high ALT value that was
recorded as an AE at follow-up. One subject in the
moxifloxacin group also had elevated AST and ALT at
follow-up.
Slight increases in blood pressure and pulse rate were
observed in subjects in the supratherapeutic THC/CBD
spray dose group, particularly on Day 1. This was
mirrored in the safety ECG results which demonstrated an
increase in ventricular rate and a decrease in QT interval
for the same group over the same time scale. In addition,
nine subjects in this group had AEs of tachycardia and
two subjects were discontinued due to intolerable AEs
which included ECG QT prolonged. There were no
apparent effects on respiratory rate and oral temperature.
Pharmacokinetic variables. Mean plasma concentrations
of THC and CBD peaked between approximately 1–
3 hours post-dose at mean concentrations of 3.1 and
1.5 ng/mL, respectively, for the 8 sprays group and 9.2
and 4.8 ng/mL for the 24/36 sprays group. In the THC/
CBD spray 24/36 sprays group, the initial peak was
followed by a second rise at approximately 6 hours post-
dose. Mean plasma levels of 11-OH-THC peaked later at
approximately 2–3 hours post-dose. The PK parameters
for THC, CBD, and 11-OH-THC are summarized in
Table 2 and mean plasma concentration versus time
curves for each analyte are presented in Figure 2. There
was a relatively high degree of inter-subject variability in
PK parameters.
Pharmacokinetic‐pharmacodynamic analysis. The slopes
for placebo-corrected change in QTcI from baseline
versus THC, 11–OH–THC, and CBD plasma concen-
trations were flat to negative. The upper bounds of
the 95% CIs for predicted change in QTcI at
average THC, 11–OH–THC, and CBD Cmax (at all
THC/CBD spray doses) were less than 10 ms
(5.821 ms). In contrast, the corresponding upper
bound for moxifloxacin was 15.61 ms. These PK-
pharmacodynamic data further support the lack of effect
of THC/CBD spray (THC, 11–OH–THC, and CBD) on
cardiac repolarization.
Discussion
This thorough QTc study at a range of doses including
supratherapeutic doses of THC/CBD spray demonstrated
292 Clinical Pharmacology in Drug Development 2(3)

Table 2. Summary of Pharmacokinetic Parameters for THC, CBD and 11–OH–THC

THC CBD 11‐OH‐THC

Cmax Tmax AUC0–t AUC0–inf Cmax Tmax AUC0–t AUC0–inf Cmax Tmax AUC0–t AUC0–inf
Parameter (ng/mL) (hour) (h  ng/mL) (h  ng/mL) (ng/mL) (hour) (h  ng/mL) (h  ng/mL) (ng/mL) (hour) (h  ng/mL) (h  ng/mL)

Median Median Median

Statistic Mean (SD) (min–max) Mean (SD) Mean (SD) Mean (SD) (min–max) Mean (SD) Mean (SD) Mean (SD) (min‐max) Mean (SD) Mean (SD)

THC/CBD 3.1 (1.64) 1.9 (0.87–23.95) 14.4 (11.29) 20.3 (11.64) 1.5 (0.78) 1.4 (0–8.45) 6.1 (5.76) 14.8 (7.87) 3.9 (2.23) 1.9 (0.87–6.48) 28.6 (23.37) 38.2 (26.90)
8 sprays
(n ¼ 60)
THC/CBD 9.2 (6.29) 2.4 (0–23.95) 63.2 (55.55) 79.3 (57.63) 4.8 (3.40) 1.5 (0–6.45) 38.9 (33.75) 60.3 (37.71) 10.0 (6.86) 2.5 (0–6.45) 90.6 (72.16) 109.4 (80.57)
24/36 sprays
(n ¼ 51)

AUC0–inf, area under the curve from time ¼ 0 to infinity; AUC0–‐t, area under the curve from time ¼ 0 to last concentration; Cmax, peak plasma
concentration; SD, standard deviation; Tmax, time to peak plasma concentration.

no significant effects on cardiac repolarization for any of
the correction methods applied. Furthermore, THC/CBD
spray had no significant effects on heart rate, PR, or QRS
interval duration or cardiac morphology. Therefore, the
results of this study reliably demonstrate that THC/CBD
spray does not significantly affect ECG parameters. The
study was valid and sensitive: moxifloxacin demonstrated
the expected increase in QTcI duration at all of the pre-
specified time points. In addition, placebo group data
demonstrated a control of background QTc variability.
The sample size used in the study was also large
compared with other similar QT/QTc studies, thus
providing a considerable amount of information regard-
ing both the cardiovascular effects of THC/CBD spray as
well as more general safety data.
The endocannabinoid system is thought to play a
key role in the control of heart rate and blood pressure,
but also contributes to pathological conditions by
affecting the heart and arterial performance through
alteration of cardiometabolic risk factors.17 The use of
cannabis on a whole has been associated with
tachycardia and transient changes in blood pressure,7–
11
and due to its effects on blood pressure and heart
frequency, its use is thought to increase the risk of heart
attacks.18 The current investigation demonstrates that
THC/CBD spray does not usually significantly affect

Figure 2. Mean (SEM) plasma concentration versus time curves for THC, 11‐OH‐THC and CBD following 8 and 24/36 sprays of THC/
CBD spray. Values below the limit of quantification are entered as zero.
Sellers et al
ECG parameters, even at supratherapeutic doses.
However, plasma levels of THC and CBD in this
study were much lower than those seen with smoked
cannabis, even at the supratherapeutic dose.
There was a relatively high degree of variability in
PK parameters for all analytes following THC/CBD
spray administration, but all data were consistent with
the results seen in previous studies19,20 and suggest that
there is little if any accumulation of cannabinoids in
plasma during chronic dosing, at least over a 5 day
dosing period.
Overall, the therapeutic dose of THC/CBD spray was
well tolerated, while supratherapeutic doses were not.
Psychiatric SAEs were reported by four subjects, all of
whom were taking 36 sprays/day. All subjects who
experienced a SAE ceased treatment and withdrew
from the study. This study was carried out in accordance
with ICH E14 Clinical Evaluation of QT/QTc Interval
Prolongation and Proarrhythmic Potential for Non-
Antiarrhythmic Drugs guidance, which suggests testing
at substantial multiples of the anticipated maximum
therapeutic exposure.1 The dose selected for this study
was expected to be close to the maximum tolerated dose
in healthy volunteers. As such, the SAE findings are not
surprising. Interestingly, the number of events of toxic
psychoses in this study suggests that THC/CBD spray was
better tolerated in this respect than the synthetic
cannabinoid, dronabinol. Numerous instances of toxic
psychoses were encountered during clinical trials of oral
dronabinol on doses as low as 16.5–20 mg THC in two of
eight cannabis-experienced subjects,21 and at even lower
doses when administered intravenously.22 Recent pub-
lications have strongly supported the ability of CBD23–25
and other phytocannabinoid and terpenoid components2
to improve the therapeutic index of THC, particularly
with respect to its psychoactive sequelae. CBD is present
in THC/CBD spray at levels similar to those of THC, and
it is reasonable to conclude that its presence may have
attenuated the psychoactivity of THC at the high doses
used in this study. Subjects taking the therapeutic dose of
eight sprays a day did not differ substantially from
subjects in the placebo group. This confirms that the
current recommended dosing schedule within the normal
clinical setting is unlikely to carry the risk of the
significant psychiatric AEs seen in this study.
Conclusions
In conclusion, this thorough QT/QTc study demonstrates
that THC/CBD spray does not significantly affect ECG
parameters, even at the maximum tolerated dose. It is not
possible to exclude an effect on the QT interval at doses
above this. This is a reassuring finding considering the
cardiovascular concerns associated with cannabis use. In
addition, THC/CBD spray is well tolerated at therapeutic
293
doses reflective of those administered in a clinical setting,
with a similar AE profile to those seen in previous clinical
studies of the drug.
Acknowledgments
With thanks to our Central ECG Laboratory eResearch
Technology, Inc. and in particular Jeff Litwin for his expert
report and advice. Thanks also to Quotient Bioresearch who
performed the bioanalysis.
Conflict of Interest Disclosure and Funding
Declaration
All authors declare: EMS, KS, CB, and MR had support from
GW Pharma Ltd for the submitted work. CFC, EBR, CGS, SW,
LW, and PD declare no support from any organization for the
submitted work. EMS, KS, CB, and MR were employees of INC
Research at the time the work was carried out, and INC was paid
by GW for conducting the research. EBR, CGS, SW, LW, and
PD are all GW Pharma Ltd employees, and as such have stock
options in GW Pharma Ltd. CFC is an employee of Otsuka
Pharmaceutical Development and Commercialization, which
has the US licence agreement for THC/CBD spray (nabiximols).
EMS, KS, CB, MR, CFC, EBR, CGS, SW, LW, and PD declare
no other relationships or activities that could appear to have
influenced the submitted work.
Clinical Trial Registration Study number: NCT01322139
URL: http://www.clinicaltrials.gov/ct2/results?term¼NCT0
1322139
References
1. International Conference on Harmonisation. E14: The
Clinical Evaluation of QT/QTc Interval Prolongation and
Proarrhythmic Potential for Non-Antiarrhythmic Drugs.
ICH Harmonised Tripartite Guideline, http://www.ich.org/
fileadmin/Public_Web_Site/ICH_Products/Guidelines/Ef-
ficacy/E14/E14_Guideline.pdf:2005 (Accessed August 23,
2012).
2. Russo EB. Taming THC: potential cannabis synergy and
phytocannabinoid–terpenoid entourage effects. Br J Phar-
macol. 2011;163(7):1344–1364.
3. Berman JS, Symonds C, Birch R. Efficacy of two cannabis
based medicinal extracts for relief of central neuropathic
pain from brachial plexus avulsion: results of a randomised
controlled trial. Pain. 2004;112:299–306.
4. Nurmikko TJ, Serpell MG, Hoggart B, Toomey PJ, Morlion
BJ, Haines D. Sativex successfully treats neuropathic pain
characterised by allodynia: a randomised, double-blind,
placebo-controlled clinical trial. Pain. 2007;133(1–3):210–
220.
5. Blake DR, Robson P, Ho M, Jubb RW, McCabe CS.
Preliminary assessment of the efficacy, tolerability and
safety of a cannabis-based medicine (Sativex) in the
treatment of pain caused by rheumatoid arthritis. Rheuma-
tology. 2006;45(1):50–52.
294
6. Kavia RBC, De Ridder D, Constantinescu CS, Stott CG,
Fowler CJ. Randomized controlled trial of Sativex to treat
detrusor overactivity in multiple sclerosis. Mult Scler.
2010;16(11):1349–1359.
7. Beaconsfield P, Grisburg J, Rainsbury R. Marihuana
smoking. Cardiovascular effects in man and possible
mechanisms. N Engl J Med. 1972;287(5):209–212.
8. Johnson S, Domino EF. Some cardiovascular effects of
marihuana smoking in normal volunteers. Clin Pharmacol
Ther. 1971;2(5):762–768.
9. Weiss JL, Watanabe AM, Lemberger L, Tamarkin NR,
Cardon PV. Cardiovascular effects of delta-9-tetrahydrocan-
nabinol in man. Clin Pharmacol Ther. 1972;13(5):671–684.
10. Paton WDM, Pertwee RG. In: Mechoulam R, ed.
Marihuana. New York and London: Academic Press;
1973;287–333
11. Paton WD. Pharmacology of marijuana. Annu Rev
Pharmacol. 1975;15:191–220.
12. Gregg JM, Campbell RL, Levin KJ, Ghia J, Elliott RA.
Cardiovascular effects of cannabinol during oral surgery.
Anesth Analg. 1976;55:203–213.
13. Fisher BA, Ghuran A, Vadamalai V, Antonios TF.
Cardiovascular complications induced by cannabis smok-
ing: a case report and review of the literature. Emerg Med J.
2005;22:679–680.
14. Sidney S. Cardiovascular consequences of marijuana use.
J Clin Pharmacol. 2002;42:64S–70S.
15. Agin MA. Assessing QT variability in healthy volunteers.
J Clin Pharmacol. 2003;43:1028.
16. Mangoni AA, Kinirons MT, Swift CG, Jackson SH. Impact
of age on QT interval and QT dispersion in healthy subjects:
a regression analysis. Age Ageing. 2003;32:326–331.
17. Montecucco F, Di Marzo V. At the heart of the matter: the
endocannabinoid system in cardiovascular function and
dysfunction. Trends Pharmacol Sci. 2012;33(6):331–340.
Clinical Pharmacology in Drug Development 2(3)
18. Grotenhermen F. The toxicology of cannabis and cannabis
prohibition. Chem Biodivers. 2007;4:1744–1769.
19. Guy GW, Robson PJ. A Phase I, double blind, three-way
crossover study to assess the pharmacokinetic profile of
Cannabis Based Medicine Extract (CBME) administered
sublingually in variant cannabinoid ratios in normal
healthy male volunteers. J Cannabis Ther. 2003; (3/4):
121–152.
20. Guy GW, Robson PJ. A Phase I, open label, four-way
crossover study to compare the pharmacokinetic profiles of
a single dose of 20 mg of a Cannabis Based Medicine
Extract (CBME) administered on 3 different areas of the
buccal mucosa and to investigate the pharmacokinetics of
CBME per oral in healthy male and female volunteers. J
Cannabis Ther. 2003; (3/4): 79–120.
21. Favrat B, Menetrey A, Augsburger M, et al. Two cases of
“cannabis acute psychosis” following the administration of
oral cannabis. BMC Psychiatry. 2005;5:17.
22. D’Souza DC, Perry E, MacDougall L, et al. The
psychotomimetic effects of intravenous delta-9-tetrahydro-
cannabinol in healthy individuals: implications for
psychosis. Neuropsychopharmacology. 2004;29(8):1558–
1572.
23. Morgan CJ, Curran HV. Effects of cannabidiol on
schizophrenia-like symptoms in people who use cannabis.
Br J Psychiatry. 2008;192(4):306–307.
24. Morgan CJ, Freeman TP, Schafer GL, Curran HV.
Cannabidiol attenuates the appetitive effects of Delta 9-
tetrahydrocannabinol in humans smoking their chosen
cannabis. Neuropsychopharmacology. 2010;35(9):1879–
1885.
25. Morgan CJ, Schafer G, Freeman TP, Curran HV. Impact of
cannabidiol on the acute memory and psychotomimetic
effects of smoked cannabis: naturalistic study. Br J
Psychiatry. 2010;197(4):285–290.